Our ESG statement data comes from identifying sub-pages on a company's website which likely relate to ESG
The recent discussion around Environmental, Social, and corporate Governance (ESG) and its importance made us curious about what we could contribute.
ESG Statements
As we're scraping each company's website, we have the ability to roughly gauge whether the website has a sub-page which relates to ESG. Pretty much every website internally links to another page. Internal links (sub-pages) are hyperlinks pointing to pages within the same domain. They serve as a navigation aid, highlighting the structure of a website and improving user experience by making it easier to find related content. In this case, we're interested in ESG content.
For now, our methodology is reasonably simple. We have a list of ESG keywords and we check whether the URL for each sub-page contains one of these words. We're particularly interested in internal links found on the homepage. A few examples of the words we're looking for include: "sustainable", "environment" and "gender". If the internal link contains one of these words, we assume the web content for that sub-page relates to ESG.
We know this methodology isn't perfect and we plan to iterate our method to improve the results but in true Data City fashion, we ship early and often. Preliminary results seem promising so it's a great way to springboard ourselves into an amazing new feature of the platform.
ESG Statements - Next steps
As most of our users know, as well as the homepage, we're scraping these sub-pages. In fact we scrape up to 75 pages per company! We can analyse the scraped text on these likely ESG sub-pages to develop a more reliable methdology. The next goal is to correctly label pages with ESG statements for each company and categorise them accordingly.
The current data live on the platform is our first step in our ESG journey hence why it's in beta on the product. We're proud to be contributing to a movement which is really important.